Modern computer systems are typically comprised of one or more memory boards and other system units, such as a central processing unit (CPU), which are coupled together by a system bus. The CPU is operable for storing data within and retrieving data from memory devices located on the memory boards. Each memory board is typically comprised of a number of dynamic read/write memory devices and may have a total memory storage capacity of many megabytes of digital data. For example, each memory board may have a total storage capacity of 16, 32, 64 or 128 megabytes of data. The range of storage location addresses which each board corresponds to are typically contiguously arranged within the address space of the system such that, for example, a first memory board may have a range of storage addresses of between zero and 16 megabytes while a second board may have a range of addresses between 16 and 32 megabytes.
As can be appreciated, a memory fault which results in a data storage error within a portion of the memory address space must be accurately identified as to the location within the address space. Related to this issue of identifying the portion of the address space having a faulty memory device is a problem related to the disabling of that portion of the memory to prevent the further occurrence of data storage errors A still further problem is introduced inasmuch as the disabling of a portion of the memory, which is preferably "mapped out" of the memory space, typically requires that a reallocation of memory addresses among the memory boards be accomplished to reflect the amount of memory which has been effectively removed from the address space of the system. Obviously, it is desirable that a minimum amount of operator or system intervention be required to reallocate the memory addresses in order to minimize or prevent system downtime and to avoid the possibility of erroneously reallocating the memory addresses.